Lire cet article en Français 
Why Tolerance Testing For Tableau
Is A Game Changer For Business Users
Tolerance Test Your Tableau Dashboards
Tolerance testing ensures business users get notified if there is an error in their Tableau dashboards when a KPI or specific data goes outside its set threshold or outside its margin of error. This testing ensures the correct data is displayed at all times and business users can trust their analytics. It also reduces team siloes because it’s business rules applied by BIA teams.
The Role Of Wiiisdom Ops
A regular request from Wiiisdom Ops users is the ability to test (assert) data against a range of values in a data warehouse. But in a production environment, data is always changing. So how can you test the data and detect anomalies?
Let’s say you have a margin of 3% today, and so you know that your margin is always between 1-5%. You won’t want to signal this dashboard as containing regressions if the margin is changing within that bracket. This margin is what the tolerance testers will allow; the total allowable error within an item. The data is updated on a daily basis and the calculation of the margin is done in the visualization, not in the database. Therefore, you don’t want the test to fail as long as the value moves between this range of 1 and 5%.
Testing this tolerance is possible in Wiiisdom Ops using formulas in Assert Data Rules. In the dashboard below, the number of new customers today is 59.
Using the formula below, Wiiisdom Ops will test if the value is above 50 and under 70.
If the test fails at any time, users will be notified that the data in their dashboard has fallen outside its set threshold.
Tolerance testing is especially good when you want to continuously monitor your dashboards to track errors before the users. This kind of test would be applicable of course during development (validating business rules) but even more in production as a regular monitoring task.
How Tolerant Are You?
Tolerance testing will avoid the risk of business users consuming and presenting dashboards with the wrong data, and therefore reduce the impact of poor decision making. If you’re interested in finding out more about this type of testing for your Tableau dashboard, get in touch with us today.
